Labov (2001) on how linguistic change happens

Saturday, November 03, 2007 11:20 AM

The conclusion (498-518) of William Labov's Principles of Linguistic Change, vol iii: Social Factors. Malden and Oxford: Blackwell. 2001.

Summary
I thought this was quite a nice summary of how linguistic change proceeds across a community. By putting together a "nonconformity principle" with a "constructive nonconformity principle", Labov ends up with a 10-stage model for the social trajectory of a linguistic change (though it apparently doesn't apply to most mergers and anything else that doesn't get symbolic value). Here's my paraphrase from Labov's pg 517-518: 1. You've got a sound, for example, that has an average pronunciation, but it also has some neighbors. Let's say that N1 is a neighbor that's further away from the average sound than N2. We can call N1 pronunciations "outliers", in this case, though, people hear the N1 as valid more often than the N2 variants. 2. You've got new language learners hearing all these pronunciations, so as time goes on, they move the average pronunciation towards N1. 3. This keeps happening for a while, irrespective of social evaluation. Thus you keep steadily shifting towards N1. 4. Younger speakers hear the N1 pronunciation as what happens when older folks aren't monitoring their speech. They know it's "deviant" from the accepted norm. 5. Younger nonconformist speakers use a higher frequency of these outliers. This comes out of the nonconformity principle: ongoing linguistic changes are emblematic of nonconformity to established social norms of appropriate behavior, and are generated in the social milieu that most consistently defies the norms. 6. Females, in particular, use more N1's than men, so these start getting heard as characteristic of female speech. 7. Male nonconformists will retreat from using N1's. 8. Upwardly mobile female nonconforming speakers spread the use of the N1's and new average pronunciation to the limits of the speech community. This is the constructive nonconformity principle, in essence: linguistic changes are generalized to the wider community by those who display the symbols of nonconformity in a larger pattern of upward social mobility. 9. You start getting social stratification, so that the overall pronunciation "develops with the highest values of the variable in communities with the highest concentration of upwardly

Definitions and examples

Nonconformity principle: ongoing linguistic changes are emblematic of nonconformity to established social norms of appropriate behavior, and are generated in the social milieu that most consistently defies the norms (2001: 516). Constructive nonconformity principle: linguistic changes are generalized to the wider community by those who display the symbols of nonconformity in a larger pattern of upward social mobility (2001: 516). Golden age principle: At some time in the past, language was in a state of perfection. (Every sound was correct and beautiful, every word and expression was proper, accurate, and appropriate. The decline has been regular and persistent and every change is a falling away rather than a return.) It's out of this that you interpret language change as nonconformity, so people reject changes in structure of language when they notice them (2001: 514). Socially motivated projection: How various factors interact to move and motivate change, how they are responsible for incrementation and transmission across generations. The idea that linguistic change is heavily constrained by the physical environment and structural factors in the language, but the forces that lead to change are largely social (2001: 498).

Open issues (mostly about race)

It's unclear what's happening cross-ethnically. In particular, things like the Northern Cities Shift are studied as white-only events. Is some linguistic change a response against blacks and others adopting "white" sounds/grammar? What's happening with Asians with regards to sound shifts (they seem not to be participating)? "Do African Americans abstain from participation in the mainstream sound changes because they do not recognize the leaders of linguistic change as reference groups, because their own system is developing in an entirely different direction, because they are isolated from these changes in their formative years, or because they do not believe that adopting these changes will gain the local rights and privileges associated with them for white speakers, or some combination of the above?" (2001: 508)

Look up
Learn about conservative pressures of functional

Socioling Page 1

with the highest concentration of upwardly mobile nonconforming speakers" (518). 10. The use of N1-outliers hits public awareness and social correction starts--this is often irregular, though. (And ineffective.)

Generalizations
"In the good majority of linguistic changes, women are a full generation ahead of men, and show a near-linear, decade-by-decade increment. Men, on the other hand, showed a more step-wise incrementation by generations for female-dominated change. A model based on the asymmetry of child-rearing fitted the Philadelphia data reasonably well" (2001: 501). "The most striking gender differentiation is found in the fact that positions in social networks are strongly correlated with women's linguistic behavior, but this is not true for men" (2001: 501). "The hypothesis is that most linguistic influence is exerted in early and middle adolescence, before the system stabilizes. The adult behavior of the leaders of linguistic change is taken as a reflection and a consequence of their behavior in their formative years" (2001: 502). "The doctrine of first effective settlement (Zelinsky 1992) limits the influence of new groups entering an established community, in asserting that the original group determines the cultural patterns for those to follow, even if these newcomers are many times the number of the original settlers" (2001: 503). "The urban newcomers' major concern is the acquisition of local rights and privileges" so their variables are targets of acquisition (2001: 505). Labov doesn't seem to like the theory of Le Page and Tabouret-Keller (1985) about social identification, he finds it simpler to posit simple interactional frequency. The social identification idea is that an individual creates systems of verbal behavior to resemble those common to the group(s) he wants (from time to time) to be identified. See Labov (2001: 505) for a quote from Le Page and Tabouret-Keller about the constraints on this. I tend to want to go down the LPTK route, though frequency is clearly crucial for transmission. A number of studies show that black and white speech patterns in Philadelphia are diverging rather than converging. "Blacks do not participate in this process [of regional dialect creation] in any large city, is a major factor in the steady and growing separation of black and white speech patterns" (2001: 507). This is not the case everywhere. When Labov played speech from black Londoners to white Londoners, none of the samples were unhesitatingly identified as black. There's more alignment along class in London, ignoring race.

Learn about conservative pressures of functional economy (2001: 500). See vol 1, chapter 20. Learn about push chains, too. Mary Haas (1944) describes dialects in Koasati having to do with men's speech and women's speech. (There's no clearly defined dialect, says Labov.) Look up Oliveira (1983) (see above) about the effect of a society's structure on language spread, re: how much social mobility is possible and where change spreads out from.

Miscellaneous
"In the course of interviewing many thousands of subjects, I and my colleagues have found older people who liked the social changes around them. They often admire the new cars, airplanes, computers, television sets, even the new music or new foods available at the supermarket. But no one has ever said, 'I really like the way young people talk today, it's so much better than the way we talked when I was growing up'" (2001: 514).

Socioling Page 2

London, ignoring race. "The most conservative Philadelphia dialect, with low front [aeo], is distinctively marked as 'white' for the black population" (2001: 508). Upward social mobility appears to be a primary characteristic for the people who lead linguistic change (2001: 509). The main resistance to dominate social norms isn't middle class hippies, but among the lower working class (2001: 510). Imitation can't lead to incrementation and children don't overshoot the mark set by their elders at least until late adolescence (2001: 512). "The first social stratification of language acquired by children is the reinterpretation of stylistic stratification on the formal/informal dimension as conforming vs. nonconforming speech" (2001: 513). Labov says the conformist/nonconformity polarity has a privileges position beyond local/nonlocal, female/male, urban/rural, modern/oldfashioned, etc. "In any society with a reasonable degree of social mobility, the CNP will tend to concentrate the leaders of linguistic change in a centrally located group--the upper working class or the lower middle class. In a society with little social mobility, they will remain in the lowest social class, and change will spread upward from that point (Oliveira 1983)" (Labov 2001: 518).

Socioling Page 3

12th class: Language change

Monday, November 05, 2007 3:53 PM

Chambers, J.K. 2002. Patterns of variation including change. In The Handbook of Language Variation and Change, ed. by J.K. Chambers, Peter Trudgill, and Natalie Schilling-Estes. Malden: Blackwell. The conclusion (498-518) of William Labov's Principles of Linguistic Change, vol iii: Social Factors. Malden and Oxford: Blackwell. 2001.

Chambers (2002)
A very important book underneath all of this is the Weinreich, Labov, and Herzog (1968). This was an important attempt to account for language change. A lot of field methods and other seeds come from here, especially for Labov. Language change is always embedded in a linguistic and social context.

Definitions and examples

Aggregate principle: When a trend is real, every additional observation gives it greater substance; when illusory, every additional observation makes it more chaotic.

Diffuse/focused: LePage's terms for the fact that the speech of older people seems more varied and less The first half of the chapter is nothing new to us, stuff predictable (diffused) than the speech of younger people (focused). (The old people might say we've already looked at (age, social class, sex). He doesn't mention ethnicity (which is crucial in the chesterfield or couch, but all the kids say couch.) Labov reading for change in America). Northern cities shift: Coffee's o is like father Phonological variables are seen as graded between Pop/lock's o is like bat classes, but grammatical variables like ain't are more Bat's /ae/ is like bet sharply graded, especially between working and Bet's e is like but middle class.

The speech community is not defined by any marked agreement in the use of language elements, so much as by participation in a set of shared norms; these norms may be observed in overt types of evaluative behavior, and by the uniformity of abstract patterns of variation which are invariant in respect to certain levels of of usage" Labov (1972: 120-121). Groups in a community mix concord and conflict (Rickford 1986). Participants are "agents in the continual construction and reproduction of the [sociolinguistic] system" (Eckert 2000: 43). Since there's some sort of consensus-without-conformity, people have latitude to express their diversity within communities. JRR says that there's certainly a dominance of the middle-class view as read by sociolinguists, even though it's the numerical minority of the population. Why does everyone in Glasgow say glottal stops are ugly but everyone still uses them?

Disagreements and quibbles

JRR doubts that middle classes in modern industrial Western societies have "no clear distinction in female mobility". I doubt this claim, too. Chambers is pushing "no change after adolescence", though this is far from clear.
JRR doesn't think the Glasgow data exhibit the same gender relationships as inner-city Detroit does.

Review
Women use fewer stigmatized and nonstandard variants than men (with class/style constant). Women may be more sensitive to prestige patterns than men. The primary correlate of change is age. Interesting claim: "In all societies, people are in most frequent and intimate daily constant with people in the same age cohortThese are the people who share reciprocal relationships, the kind that carry the most weight both socially and linguistically" (2002: 366). Abrupt change is rare (because people similar to one another in age speak similar to one another, even with rapid change.

Labov (2001)

From Labov (1972: 161-162), if you're looking at language change in progress, you have to solve three problems: a. Transition: How do you find the route by which one stage of a linguistic change evolved from an Miscellaneous earlier stage? Look at children being raised by two mothers b. Embedding: How do you find the continuous

Socioling Page 4

Look at children being raised by two mothers b. Embedding: How do you find the continuous and two fathers (around 2-7?) see how they're matrix of social and linguistic behavior in which patterning. the change is carried? c. Evaluation: How do you find the subjective (or Everyone who studies change likes to come up with latent) correlates of the objective (or manifest) principles, says JRR. (Golden age principle, aggregate changes that have been observed? You go for principle, etc.) measuring attitudes/aspirations/reactions of consultants. Everett Rogers' Diffusion of Innovations sounds interesting. His principles in (2001) bear some resemblance to the (1972) principles, but they've evolved (see my earlier, separate summary, which includes these).

Who are the leaders of linguistic change? Well, the people in the middle of the social class skill. People with mobility. There's a missing piece (acknowledged as such) when it comes to ethnicity and change. There's evidence that African-Americans don't participate in many of the white changes. On the other hand, JRR says the "all project" showed no difference in race, just age.
I like the nonconformity hypothesis, but I wonder about it relative to filler-like. There's a motivation in the linguistic marketplace to not say "like" all the time (people really won't hire you), but many people seem incapable of controlling it. What is the role of conformity here? Consciousness?

Socioling Page 5